Technical SEO for AI Search · Explainer
Do AI crawlers render JavaScript? What it means for your site
Part of the guide Technical SEO for AI search: the 2026 checklist, in order
Most AI crawlers do not run JavaScript. Vercel's analysis of crawler traffic (published 17 December 2024) found that "none of the major AI crawlers currently render JavaScript", naming GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot and PerplexityBot. Googlebot does render JavaScript, after a queue. So if your text only appears after scripts run, Google may read it while ChatGPT, Claude and Perplexity see an empty page.
Which crawlers render JavaScript and which do not?
Googlebot and Applebot render; the AI assistants' crawlers, in the best public test we found, do not. The table combines Google's own documentation with Vercel's measurements of traffic across its network.
Rendering is the third check in our technical SEO for AI search checklist, right after bot access and robots.txt.
| Crawler | Runs JavaScript? | Source and date |
|---|---|---|
| Googlebot (Search, AI Overviews, AI Mode) | Yes, in a separate rendering phase | Google JavaScript SEO basics, updated 4 March 2026 |
| Applebot | Yes, "through a browser-based crawler" | Vercel, 17 December 2024 |
| GPTBot, OAI-SearchBot, ChatGPT-User | No | Vercel, 17 December 2024 |
| ClaudeBot | No | Vercel, 17 December 2024 |
| PerplexityBot | No | Vercel, 17 December 2024 |
| CCBot (Common Crawl) | No, "does not render pages" | Vercel, 17 December 2024 |
Vercel also found that ChatGPT's and Claude's crawlers do download JavaScript files (11.50% and 23.84% of their requests) but "don't execute them". Downloading a script is not running it.
Two caveats. The Vercel data is almost two years old and vendors can change crawlers without notice; none of OpenAI's, Anthropic's or Perplexity's bot pages we read on 16 November 2026 say they render JavaScript, so we have no newer primary source either way. And Claude-SearchBot and Claude-User were not named in the study. Until a vendor documents otherwise, assume a non Google bot reads only the HTML your server sends.
How does Google handle JavaScript?
Google processes JavaScript pages in three phases: crawling, rendering and indexing. Its documentation says "Googlebot queues pages for both crawling and rendering" and that a page "may stay on this queue for a few seconds, but it can take longer than that." Every 200 page is queued for rendering unless a robots meta tag or header says not to index it.
So Google will usually see client rendered text, but later and with more ways to fail: a blocked script file, an API call that times out, a component that needs a click. Google itself recommends against relying on it: "server-side or pre-rendering is still a great idea because it makes your website faster for users and crawlers, and not all bots can run JavaScript."
If Google keeps crawling a client rendered page without indexing it, a thin first render is one cause worth ruling out; see how to fix crawled, currently not indexed.
How do you see what an AI crawler sees?
Fetch the page without a browser and look for your body text. This is the whole test:
curl -s -A "OAI-SearchBot" https://www.example.com/pricing | grep -c "Your plan includes"
Replace the phrase with a sentence from the middle of the page. A count of 0 means the sentence is not in the server HTML. To see what is there instead:
curl -s https://www.example.com/pricing | perl -0pe 's/<(script|style)\b.*?<\/\1>//gs; s/<[^>]+>/ /g' | tr -s '[:space:]' ' ' | head -c 1500
That strips tags and prints the visible text the server sent. If it is mostly menu items and a cookie notice, the content is client rendered. For Google's view, use URL Inspection in Search Console: "View crawled page" shows the HTML Googlebot got, and the live test shows a screenshot. The full multi bot version of the curl check is in test AI crawler access with curl.
What usually goes missing without JavaScript?
It is rarely the whole page. More often one part that matters for answers is built in the browser. Check these first:
- Prices and plan tables loaded from a pricing API after page load. These are exactly what AI answers quote.
- Reviews and ratings widgets injected by a third party script.
- Tabs and accordions whose content is fetched when clicked, rather than present in the HTML and hidden with CSS. Hidden but present is fine for crawlers that read HTML; fetched on click is invisible.
- JSON-LD added by a tag manager. A crawler that does not run scripts never sees structured data injected that way. Put it in the server HTML.
- Infinite scroll and "load more" on blog indexes, which hides older posts from any crawler that does not scroll. Give them paginated URLs.
- Internal links rendered by JavaScript, which stops non rendering crawlers finding your other pages.
How do you fix it?
Send the content in the first HTML response. You have four ways to do that, from most to least robust:
| Approach | What it means | Good for | Effort |
|---|---|---|---|
| Static generation | HTML built at deploy time | Blogs, docs, marketing pages | Low with most frameworks |
| Server side rendering | HTML built per request | Pages with live data, such as pricing | Medium |
| Prerendering service | A service renders the page and serves the result to bots | Legacy single page apps you cannot rebuild | Medium; must keep bot and user content identical |
| Client rendering with critical text in HTML | App stays client side but key text ships in the initial HTML | Apps where only some pages need to be found | Varies |
Frameworks such as Next.js, Nuxt, Astro and SvelteKit can render pages on the server or at build time; a plain single page app built with React, Vue or Angular renders in the browser unless you add one of those modes. Content management systems such as WordPress, Ghost and Webflow send post text in the HTML by default; the risk on those platforms comes from plugins and embeds, not the post body.
Whatever you choose, show bots the same content users see. Google's spam policies define cloaking as "presenting different content to users and search engines with the intent to manipulate search rankings and mislead users".
What about the rest of the technical list?
Rendering is the second door after access. Once your text is in the HTML, the remaining work is canonical tags, an accurate sitemap and structured data that ships in the HTML. The order and checks are in our technical SEO for AI search checklist, and bot access rules are in the robots.txt guide for AI crawlers.
Articles that Postbing publishes to WordPress, Webflow, Ghost or Shopify arrive as finished HTML in the CMS, including the FAQ and JSON-LD, so their text does not depend on scripts. Your theme and plugins still decide what else loads. The integrations page lists what each connector writes.
FAQ
Does ChatGPT's crawler run JavaScript?
In Vercel's December 2024 analysis, GPTBot, OAI-SearchBot and ChatGPT-User did not execute JavaScript, although GPTBot did download some script files. OpenAI's own bot page does not say either way.
Does Googlebot render JavaScript?
Yes. Google crawls, queues the page for rendering, renders it and then indexes it. Google still recommends server side or pre-rendering because not all bots can run JavaScript.
Is my React site invisible to AI search?
Only if the text is rendered in the browser. A React site using server rendering or static generation sends the text in the HTML. Test with curl and look for a sentence from the page.
Can AI crawlers read JSON-LD added by Google Tag Manager?
Crawlers that do not run JavaScript cannot see anything a tag manager injects, including JSON-LD. Put structured data in the server HTML.
Is prerendering for bots cloaking?
Not if bots and users get the same content. Serving bots different text from what users see violates Google's spam policies.
Sources
- Vercel, The rise of the AI crawler: which crawlers render JavaScript, script download shares, published 17 December 2024, accessed 16 November 2026
- Google Search Central, JavaScript SEO basics: crawl, render, index phases, rendering queue, server side rendering advice, accessed 16 November 2026
- Google Search Console Help, URL Inspection tool: View crawled page, live test screenshot, accessed 16 November 2026
- Google Search Central, spam policies: definition of cloaking, accessed 16 November 2026
Related reading
Last updated: November 16, 2026. Spotted an error? Email support@postbing.com and we will fix it. Editorial policy.
Try Postbing free
Postbing reads your site, plans a month of topics from what your customers search for, writes in your voice and publishes to WordPress, Shopify, Webflow or Ghost. Your first article is free.
Try Postbing freeNo card needed. No link exchanges, ever.